DocumentCode :
3088990
Title :
A comparative study of partial discharge by classification´s kind
Author :
Jaramillo-Vacio, R. ; Ochoa-Zezzatti, A. ; Rios-Lira, A. ; Cordero, D.
Author_Institution :
Comision Fed. de Electr.-Lab. de Pruebas a Equipos y Mater. (LAPEM), Mexico
fYear :
2012
fDate :
4-7 Dec. 2012
Firstpage :
34
Lastpage :
40
Abstract :
This paper presents a comparison of competitive learning algorithms for Self Organizing Map (SOM). The competitive learning algorithms showed to self organizing map algorithm are winner-takes-all, Frequency Sensitive Competitive Learning and Rival Penalized Competitive Learning. The result shows the performance in classification of partial discharge on power cables using SOM.
Keywords :
partial discharges; pattern classification; power cables; power engineering computing; self-organising feature maps; unsupervised learning; SOM algorithm; frequency sensitive competitive learning algorithm; partial discharge classification; power cables; rival penalized competitive learning algorithm; self-organizing map algorithm; winner-takes-all competitive learning algorithm; Algorithm design and analysis; Discharges (electric); Insulation; Partial discharges; Power capacitors; Training; Vectors;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Hybrid Intelligent Systems (HIS), 2012 12th International Conference on
Conference_Location :
Pune
Print_ISBN :
978-1-4673-5114-0
Type :
conf
DOI :
10.1109/HIS.2012.6421305
Filename :
6421305
Link To Document :
بازگشت